Hand bone radiograph image segmentation with ROI merging
ACC'11/MMACTEE'11 Proceedings of the 13th IASME/WSEAS international conference on Mathematical Methods and Computational Techniques in Electrical Engineering conference on Applied Computing
| Hi-index | 0.00 |
Many applications of geometry processing and computer vision rely on geometric properties of curves, particularly on their curvature. So the estimation of digital curve’s curvature is important. An algorithm of curvature estimation is proposed in this paper, which is based on the recognition of discrete straight segments. For each point of a digital curve, its curvature is estimated by searching the discrete tangents of this point and of its neighboring points. To test the effectiveness and the precision of this algorithm, it’s compared with Hermann algorithm by a lot of experiments conducted on the digital circles in a wide range of resolution. For each curve, the maximum and the mean of errors are recorded. Experimental results show that the new method has an overall good performance.